14.

13 Lecture 14

Xavier Gabaix

April 1, 2004

Decision utility
Economists use the word utility to describe the preferences that rationalize observed choices.

Kahneman calls these revealed preferences decision utility.

These are the preferences that rationalize decisions.
These are the preferences that coincide with wanting and choosing.
For an addict the decision utility of drug consumption exceeds the decision
utility of quitting.

Experienced utility
Kahneman also measures the hedonic consequences of choices.
He calls these actual hedonic experiences, experienced utility.
These are the preferences that coincide with doing.
This is how Jeremy Bentham (1748-1832) conceived of utility (pleasure
and pain)
How do you measure hedonic experiences (e.g., well-being)?
How do you discount these experiences over time?

Why might decision utility and experienced utility fail to coincide? A couple
examples:

cognitively hard decision to make, e.g.,

choosing a job
planning for retirement

 poor forecasts of preference dynamics, e.g.,

failures to anticipate adaptation = "psychological immune system" cf
Gilbert et al (paraplegics, lottery winners, junior faculty turned down
for tenure)
Happiness

realized

Adaptation

forecasted
Accident

Time

failures to anticipate visceral eects (Loewensteins aggression study)

 inaccurate memories of past hedonic experiences (this is where well focus

today)

much of this course is about occasional disconnects between decision utility

and experienced utility

Remembered utility
our memory of a hedonic experience
remembered utility exhibits duration neglect
remembered utility follows peak-end rule: global retrospective evaluations
are well predicted by an average of the peak aective response recorded
during an episode and the end value recorded just before the termination
of an episode

Examples of studies:

colonoscopy
plotless films of pleasant/unpleasant subjects (such as low-level flying over
an African landscape or of amputation)

immersion of one hand in cold water (cold-pressor)

aversive sounds of varying loudness and duration
shocked rats

Colonoscopy (Katz, Redelmeier, and Kahneman)

control group: regular colonoscopy
treatment group: procedure lengthened by one minute with colonoscope
inside the body cavity but stationary (nature of condition was not explained
to the subjects)

treatment group had significantly better memories of the overall experience

how could these insights be used practically?

10

Pain Intensity

Pain Intensity

10

6
4
2

6
4
2
0

0
0

10

20

10

Time (minutes)
Patient A

Patient B

Pain Intensity Reported by Two Colonoscopy Patients,

adapted from Kahneman, et al. "Back to Bentham?
Explorations of Experienced Utility. " The Quarterly
Journal of Economics 112, no. 2 (May 1997).

20

Cold Pressor (Schreiber & Kahneman)

short trial: hand in 14 degree water (60 sec)
long trial
hand in 14 degree water (60 sec)
temp rises to 15 degrees (30 sec)
65% of subjects chose to repeat the long trial (so decision utility 6= experienced utility)
result replicated with aversive noise

Comparison of Happiness across countries and

time
Source: Frey and Stutzer What Can Economists Learn from Happiness
Research?, JEL 2002.

A simple way to measure happiness: Ask

World Values Survey asks, on a scale of 1 to 10: All things considered,
how satisfied are you with your life as a whole these days?
US: General Social Surveys (scale of 1 to 3) Taken all together, how
would you say things are these dayswould you say that you are very
happy, pretty happy, or not too happy?

15000

3.5
Real GDP
per capita
3

9000
Life satisfaction

2.5

6000
2
3000

1.5

1958

1962

1966

1970

1974
Year

1978

1982

1986

1990

Satisfaction with Life and Income Per Capita in Japan between 1958 and 1991,
adapted from Frey, et al. "What Can Economists Learn from Happiness Research?"
Journal of Economic Literature 40, no. 2 (June 2002).

Average life satisfaction

Real GDP per capita in constant $

12000

Average Happiness

2.5

2.3

1972 - 74
1994 - 96
Trend 1972 - 74
Trend 1994 - 96

2.1

1.9

10000

20000

30000

40000

50000

60000

70000

Equivalence Income in 1996 US$

Happiness and Equivalence Income in the United States, adapted from Frey, et al.
"What Can Economists Learn from Happiness Research?" Journal of Economic
Literature 40, no. 2 (June 2002).

 Japan: no visible correlation between GDP and happiness. Why?

adaptation u = u( c ct )
t1

material success is not happiness (not robust)

interviewees interpret the question as "relative happiness" (the average
answer to "how tall are you?" doesnt change over the years even if
the average height has increased a lot)
expectation wise u = u( E ct [c ] )
t1 t
happiness measured from 1 to 10 whereas in principle it could be unbounded